Get alerts on this situation

We’ll email you as it develops, and you can follow the whole thread from day one.

Unsubscribe anytime.

[SITUATION] · [ACTIVE]

2 clusters · 9 sources · 7 days · First seen 9 days ago · Last updated 2 days ago

AI-driven hurricane forecasting advancements

Entities: Institute of Atmospheric Physics, Chinese Academy of Sciences · FuXi · Google DeepMind · WeatherNext · Mike Brennan

Overview

In late July 2026, Chinese researchers unveiled an AI‑based forecasting system that integrates the FuXi weather model with realistic ensemble generation to markedly improve hurricane track predictions. Tests on dozens of tropical cyclones showed the system outperformed leading global ensemble models and could be adapted to other hazardous weather events.

A week later, DeepMind announced WeatherNext, an AI model that extends forecast capability by predicting both intensity and track a full day earlier than existing models. In a trial on Hurricane Melissa, WeatherNext correctly anticipated rapid intensification and landfall, giving the U.S. National Hurricane Center extra time for warnings and evacuations. The model leverages low‑resolution data to generate thousands of scenarios and involves collaboration with multiple national weather agencies.

On 5 August 2026 DeepMind and Google Research released the WeatherNext code and model weights as open‑source software. The system can generate a 15‑day forecast in under a minute on a TPU and produces unified track‑and‑intensity predictions up to a day ahead of operational models. Early adopters such as the U.S. National Hurricane Center, the Cooperative Institute for Research in the Atmosphere and the UK Met Office have begun testing the tool, noting that the extra lead time could improve preparedness and reduce casualties.

Together, these developments illustrate a rapid international push to harness artificial intelligence for more accurate and earlier hurricane forecasts, enhancing early‑warning systems and disaster mitigation worldwide.

Claims

What the coverage asserts, and how well corroborated each claim is across sources.

[● 5 SOURCES] WeatherNext can predict cyclone track and intensity up to a day earlier than existing operational models. (DeepMind AI model)
[● 5 SOURCES] The model was trained on about 20 TB of global atmospheric data and historic storm records. (Google DeepMind research team)
[● 4 SOURCES] WeatherNext can generate a 15‑day forecast in less than a minute on a TPU. (Google DeepMind)
[● 4 SOURCES] The project involved collaboration with Google DeepMind, Google Research, the U.S. National Hurricane Center, the Cooperative Institute for Research in the Atmosphere and the UK Met Office. (Google DeepMind)
[● 3 SOURCES] WeatherNext’s code and model weights have been released as open‑source on GitHub. (Google DeepMind)
[● 3 SOURCES] In tests, WeatherNext correctly forecast the rapid intensification and landfall of Hurricane Melissa, giving authorities extra warning time. (Google DeepMind team)
[● 2 SOURCES] Mike Brennan, director of the U.S. National Hurricane Center, said an extra day of warning could improve evacuations and reduce risk to communities. (Mike Brennan)
[● 2 SOURCES] The model’s forecast skill represents roughly a decade of progress condensed into a single AI system. (Google DeepMind)

Timeline

2 days ago

[TECHNOLOGY] 9 sources
DeepMind's WeatherNext AI adds a day to cyclone forecasts
DeepMind's WeatherNext AI predicts cyclone tracks and intensity a day earlier, offers 15‑day forecasts in minutes, is open‑source, and has been tested on Hurricane Melissa, improving early warnings.
9 days ago

[TECHNOLOGY] 2 sources
China's AI System Boosts Hurricane Track Forecast Accuracy
Chinese scientists unveiled an AI system that outperforms global models in forecasting hurricane tracks, speed and direction, promising better early warnings.

Sources

1001web.fr · archynetys.com · areamobile.de · bright.nl · dev.to · geeknetic.es · ilcorrierino.com · sg.com.mx · worldcatholicnews.com

This summary has been updated 1 time: see revision history